Work

Open SourceGitHub, npm, PyPI, and public packages.

Depth

JourneyCareer timeline, roles, and education.AchievementsScholarships, rankings, and milestones.

Writing

PublicationsResearch-facing work and technical reports.ProfilesAll key public links in one place.

Professional Journey

The work changed as the questions got harder

I started by making interfaces work. Each role widened the problem: APIs, cloud systems, team decisions, AI workflows, and the responsibility to leave a codebase clearer than I found it.

5+ years
Experience
7 + 3
Roles & education
45
Capabilities mapped
10
Working domains
Frontend beginningsFull-stack ownershipProduction systemsFinancial engineeringResearch questionsProduct engineeringAI systemsCloud architectureData workflowsTechnical leadership
Frontend beginningsFull-stack ownershipProduction systemsFinancial engineeringResearch questionsProduct engineeringAI systemsCloud architectureData workflowsTechnical leadership

  1. 4+

    Professional years

  2. 7

    Work roles

  3. 5

    Work contexts

  4. 2

    Education paths

  5. 4+ yrs

    TypeScript and web delivery

  6. 300M+

    Events handled per tenant

  7. 12×

    Demonstrated query improvement

  8. 8

    Public developer tools

The actual timeline

Roles, detours, and decisions that added up

Nine dated records of where I worked or studied, what I was responsible for, and which tools travelled with me.

·Karachi, Pakistan
ExperienceVentureDive

Senior Software Engineer

Building production AI-enabled full-stack applications, with a focus on keeping LLM reasoning, deterministic business logic, and application architecture cleanly separated as AI-assisted development scales across the team.

  • Own features end-to-end across frontend, backend, AI integration, and deployment rather than treating AI as a bolt-on layer.
  • Apply prompt and context engineering plus spec-driven development to make AI-assisted output predictable enough to ship.
  • Review AI-generated implementations against real architecture and maintainability constraints before they reach production.
Next.jsTypeScriptAWSLLM IntegrationAgentic Engineering
·Troutdale, Oregon, USA (Remote)
ExperienceDunzo

Business Automation & AI Engineer

Building agentic workflows for service businesses — lead intake, estimate generation, job routing — wired into GoHighLevel, HouseCall Pro, Jobber, and ServiceTitan, with retrieval and validation steps before any automated action runs.

  • Designed workflows around verifiable completion states so automation gets judged on outcomes, not plausible-looking output.
  • Synchronized business and operational data across systems into machine-readable entities for retrieval and automation.
  • Shipped an AI-assisted advertising automation system that generates campaign assets and uses performance data to inform decisions.
n8nGoHighLevelLLM WorkflowsCRM IntegrationGTM Automation
·Online
EducationWorldQuant University

MSc in Financial Engineering

Structured study in financial modeling, econometrics, and AI-driven decision systems as part of a stronger move into fintech and graduate-oriented work.

Financial EngineeringEconometricsModelingPython
·Online
EducationValar Institute

MBA in Leadership and Management

Graduate study in leadership and management on a 75% merit scholarship, taken alongside full-time engineering work.

LeadershipManagementStrategy
·Karachi, Pakistan
ExperienceDigital Dividend Global

Software Engineer Level II

Led platform modernization, AI-enabled delivery, and architecture improvements across frontend, backend, cloud workflows, and internal developer systems.

  • Architected a custom agent framework chaining tool execution across ETL, documents, Teams, n8n, Zapier, and Outlook.
  • Modernized a legacy codebase into a clearer monorepo architecture with stronger workflows.
  • Optimized AWS Athena workloads across 300M+ events per tenant for roughly 12x faster queries.
TypeScriptReactNext.jsAWSTerraformYarn Workspaces
·Karachi, Pakistan
ExperienceDigital Dividend Global

Software Engineer Level I

Worked across executive CMS systems, payment integrations, reporting, cloud deployment improvements, and platform performance optimization.

  • Built and improved CMS-backed platforms with payments, admin workflows, and reporting.
  • Integrated services such as Tamara, HyperPay, Apple Pay, Securiti.AI, and ZATCA.
  • Improved code quality through better indexing, query work, and deployment refinement.
JavaScriptNode.jsMongoDB.NETReactTerraform
·Karachi, Pakistan
ExperienceStar Marketing Pvt. Ltd.

Frontend Engineer

Rebuilt the company site for speed, SEO, and broader content delivery, then handled cost-aware platform work and a CMS migration across the same systems.

  • Reduced initial load time by roughly 5x while supporting substantially more content.
  • Improved city-level SEO visibility for global-facing pages.
  • Saved infrastructure cost across ElastiCache, App Runner, and Amplify.
  • Supported migration from Payload CMS and MongoDB to Strapi and PostgreSQL.
Next.jsJavaScriptSSRSEOStrapiPostgreSQLAWS
·Karachi, Pakistan
ExperienceDivine Virtuality

Software Engineer

Worked across responsive interfaces, API integration, cloud deployments, and cross-platform product delivery.

  • Built reactive frontend experiences with TypeScript, Next.js, and component libraries.
  • Supported reliable deployments and uptime through AWS and Azure-oriented infrastructure work.
TypeScriptReactNext.jsReact NativeAWSAzure
·Antalya, Turkiye
ExperienceCivar.com

Frontend Developer

Contributed to a dynamic e-commerce platform with reusable components, responsive UI work, and product-focused frontend delivery.

  • Built scalable UI components for a growing e-commerce platform.
  • Worked with Next.js, GraphQL, NextAuth, and design handoff processes.
ReactNext.jsGraphQLNextAuthJavaScript
·Karachi, Pakistan
EducationBahria University

Bachelor of Science in Computer Science

Completed BSCS with Cum Laude distinction and a 3.76 / 4.0 CGPA, building a strong foundation in algorithms, systems, AI, and distributed computing.

AlgorithmsData StructuresAINeural NetworksDistributed Systems

What changed between roles

Growth looked more like widening responsibility than a ladder

The meaningful transitions were not title changes. They were the moments when I stopped treating the visible task as the whole problem and started owning the next constraint too.

Own the next constraint, not only the assigned ticket

Translate between product intent, interface detail, and system behavior

Document and teach whatever becomes repeatable

The through-lines

Three changes explain the shape of the journey

The stack expanded, but the more important shift was how I framed and owned the work.

From interfaces to systems

Frontend work taught me to care about the last mile. Backend and cloud work taught me how many decisions determine whether that last mile stays reliable.

From implementation to judgment

The job became less about knowing a framework and more about choosing what to simplify, what to measure, and where complexity was actually worth its cost.

From delivery to research questions

Financial engineering and applied AI widened the questions again: how evidence, uncertainty, and incentives should shape software decisions.

Capability map

Where the depth comes from

These bands are directional, not a certification score. Open a capability to see the kind of work behind it.

agentic

05 skills
Strong working depth
AI Agents and Orchestration
2+ yr

Agent loops, tool calling, completion criteria, and multi-agent handoffs built to survive real inputs.

Claude Code, Codex and Cursor
2+ yr

Daily AI-assisted engineering at team scale, including the review discipline that keeps it honest.

Spec-Driven Development
2+ yr

Writing the spec before the prompt so AI-assisted output stays reviewable and reproducible.

Agent Evaluation and Guardrails
2+ yr

Evals, failure taxonomies, and guardrails — the part that decides whether an agent is production-safe.

Custom Agentic Frameworks
2+ yr

Built in-house where off-the-shelf orchestration hid too much of the control flow to debug.

ai

09 skills
Strong working depth
Prompt and Context Engineering
3+ yr

Structured outputs, meta-prompting, and context budgeting so model behaviour stays predictable enough to ship.

Python
3+ yr

Automation, data work, AI integration, scripts, FastAPI services, and modeling-oriented workflows.

RAG and Semantic Retrieval
2+ yr

Chunking, embedding, and re-ranking strategies tuned against real corpora rather than toy datasets.

OpenAI and Agent Workflows
2+ yr

Assistants, prompt design, workflow orchestration, and production-minded AI integration.

Vector Stores
2+ yr

Index design and recall/latency tuning for retrieval that has to answer in under a second.

Pandas
3+ yr

Data preparation and analysis for modelling work and pipeline validation.

Gemini
2+ yr

Multimodal and long-context work, mostly in hackathon and prototype settings that later hardened.

LangChain
2+ yr

Practical use in retrieval and agent-like workflows rather than abstract demos.

PyTorch
2+ yr

Model training and experimentation behind the published research on image classification and action recognition.

blockchain

01 skills
Applied experience
Solidity and Web3 Product Work
2+ yr

Smart contracts, dApp concepts, wallet integration, and token-aware prototypes.

cloud

04 skills
Strong working depth
AWS
3+ yr

Strongest cloud platform across Lambda, S3, Athena, DynamoDB, Kinesis, and supporting services.

Docker
3+ yr

Container-based local development, service packaging, and platform workflows.

Azure DevOps and Datadog
3+ yr

Pipelines and production observability for the services I ship.

Terraform
2+ yr

Used for infrastructure as code, deployment consistency, and operational cleanup.

data

07 skills
Strong working depth
Athena and Query Optimization
2+ yr

High-leverage work around analytics performance, throughput, and large-scale query behavior.

MongoDB
4+ yr

Document modeling, application integration, and real-world product delivery.

Data Pipelines
3+ yr

Event handling, ETL patterns, observability pipelines, and analytics-oriented architecture.

PostgreSQL
3+ yr

Schema design, migrations, relational modeling, and production app integration.

DynamoDB
3+ yr

Access-pattern-first modelling for high-write observability workloads.

Kafka and Kinesis
3+ yr

Event streaming at the volumes where batch stops being an option.

Redis
2+ yr

Caching and rate limiting in front of expensive model and query calls.

fintech

03 skills
Applied experience
Financial Engineering Foundations
1+ yr

Quant-oriented study shaping how I think about systems, risk, and decision quality.

Quantitative Modelling
1+ yr

Econometrics, pricing, and portfolio risk from the WorldQuant MSc, applied to trading-adjacent systems.

Algorithmic Trading Tooling
2+ yr

TradingView, Alpaca, MQL5, and backtesting — a secondary track, kept honest about its scope.

geo

04 skills
Strong working depth
Generative Engine Optimization (GEO)
1+ yr

Making a business legible to ChatGPT, Claude, Perplexity, and AI Overviews rather than only to Google's blue links.

Machine-Readable Business Data
1+ yr

Structured data, llms.txt, and entity modelling so crawlers and answer engines read the same facts.

Citation and Brand-Mention Monitoring
1+ yr

Tracking where and how an entity gets cited across answer engines, and closing the gaps that show up.

Entity Discoverability
1+ yr

Citation-gap analysis and semantic search positioning for brands that are invisible to AI search.

leadership

01 skills
Strong working depth
Technical Leadership and Mentorship
2+ yr

Code review standards, mentoring junior developers, and helping teams make sharper technical choices.

other

03 skills
Strong working depth
GTM and RevOps Automation
2+ yr

Lead routing, CRM hygiene, outbound sequencing, and pipeline reporting built as agentic workflows instead of manual busywork.

n8n and Zapier Workflow Design
2+ yr

Multi-step automation across CRMs, forms, and internal tools, wired to real system-of-record data rather than one-off scripts.

CRM Integration
2+ yr

GoHighLevel, HouseCall Pro, Jobber, and ServiceTitan integrations for lead intake, estimates, and job routing.

web

08 skills
Core strength
TypeScript
4+ yr

Primary language for frontend, backend, tooling, and product architecture work.

JavaScript
4+ yr

Strong production experience across React, Node.js, browser APIs, and cross-platform delivery.

Next.js
4+ yr

SSR, SEO, App Router, platform modernization, and high-performance delivery.

React
4+ yr

Component systems, app architecture, performance optimization, and product-facing UI delivery.

Node.js
4+ yr

APIs, backend systems, services, automation, and event-oriented architecture.

C#/.NET Core
3+ yr

Backend microservices for enterprise-scale event analytics at AppNavi.

FastAPI
2+ yr

Python services fronting AI workloads where request shape and latency both matter.

React Native
3+ yr

Cross-platform delivery for the CarHub marketplace, shared with the web codebase.

How the pieces connect

A product engineer first, with depth where the system needs it

The useful part of breadth is being able to follow a problem across the interface, API, data model, infrastructure, and operating workflow without losing the product decision that started it.

Start from the user and business constraint

Go deep in the bottleneck instead of decorating every layer

Leave behind a system the next engineer can understand

How I learn

Depth comes from pressure, reflection, and reuse

A technology becomes a real capability only after it survives constraints outside a tutorial.

Ship it under a real constraint

Use the tool where latency, permissions, reliability, cost, or team ownership can expose weak assumptions.

Measure the trade-off

Compare the result with the simpler option and keep evidence of what actually improved.

Teach or package the lesson

Turn durable patterns into reviews, writing, reusable modules, or clearer team conventions.

Technology ecosystem

Tools are supporting characters

The stack changes with the work. These are the tools that currently recur across shipped systems, experiments, and public packages.

  • AI Agents and Orchestration
  • AWS
  • Agent Evaluation and Guardrails
  • Algorithmic Trading Tooling
  • Athena and Query Optimization
  • Azure DevOps and Datadog
  • C#/.NET Core
  • CRM Integration
  • Citation and Brand-Mention Monitoring
  • Claude Code, Codex and Cursor
  • Custom Agentic Frameworks
  • Data Pipelines
  • Docker
  • DynamoDB
  • Entity Discoverability
  • FastAPI
  • Financial Engineering Foundations
  • GTM and RevOps Automation
  • Gemini
  • Generative Engine Optimization (GEO)
  • JavaScript
  • Kafka and Kinesis
  • LangChain
  • Machine-Readable Business Data
  • MongoDB
  • Next.js
  • Node.js
  • OpenAI and Agent Workflows
  • Pandas
  • PostgreSQL
  • Prompt and Context Engineering
  • PyTorch
  • Python
  • Quantitative Modelling
  • RAG and Semantic Retrieval
  • React
  • React Native
  • Redis
  • Solidity and Web3 Product Work
  • Spec-Driven Development
  • Technical Leadership and Mentorship
  • Terraform
  • TypeScript
  • Vector Stores
  • n8n and Zapier Workflow Design
  • LLM Integration
  • Agentic Engineering
  • n8n
  • GoHighLevel
  • LLM Workflows
  • GTM Automation
  • Financial Engineering
  • Econometrics
  • Modeling
  • Leadership
  • Management
  • Strategy
  • Yarn Workspaces
  • .NET
  • SSR
  • SEO
  • Strapi
  • Azure
  • GraphQL
  • NextAuth
  • Algorithms
  • Data Structures
  • AI
  • Neural Networks
  • Distributed Systems

Next Steps

Continue exploring

What Skills Do You Need to Build AI Agents in 2026?

Building production AI agents requires more than prompting. Discover the full skill stack: from Python async patterns to evaluation frameworks.

Navigate

Why Every Developer Should Learn Basic Self-Hosting

This isn't a pitch to move your production app off Vercel. It's an argument that not knowing what a reverse proxy, a process manager, or a TLS handshake actually does puts a ceiling on how good a debugger you'll ever be — and a $5 box you're allowed to break is enough to fix it.

Navigate

Should Software Engineers Become AI Engineers?

Mostly no — and the reason is in the data people cite to argue yes. The Stanford AI Index finds the fastest-growing AI skills are deployment ones: AWS, scalability, workflow management. The market is short of engineers who can ship these systems, not people who understand them.

Navigate

Prompt Engineering Is a Floor, Not a Career

Learn it — it takes about a week and it genuinely helps. Then stop, because every interface improvement is deliberately designed to make your prompting skill worthless, and the vendors are quite open about that being the goal.

Navigate

Where to go next

What the roles produced

Where these skills get used

The rest of the record